I would suggest that something related but subtler is responsible for the apparent genetic contribution to political and social beliefs. People are persuaded by different types of evidence, and this could quite easily have a brain-structural component. This is fraught with peril, of course, because everyone will claim that their side is persuaded by strictly logical argument, while their opponents are responding to base emotional appeals. But the standard list of persuaders, including appeal to authority, social pressure, anecdote, etc, are not mutually exclusive. All sides at least attempt all methods.
For any large social issue, reasons for being on one side or another will not be homogenous across the group. There may, however, be strong tendencies, and these would be self-reinforcing. People who respond to appeal-to-authority arguments might well cluster together, and keep feeding each other appeal-to-authority arguments, reinforcing the original persuader with strong social ties.
This could also change over time. The group of people in favor of the death penalty in 1950 might be somewhat different than the group in favor in 2000, as the types of persuasion, and the predominance of the belief in certain social or ethnic groups might gradually change. It is certainly obvious anecdotally that certain professions tend strongly to particular political beliefs. Social workers are overwhelmingly liberal. Is this only because liberals go into social work, or is there some deeper affinity, even at the brain level, of how both groups perceive change and how the world works? Interestingly, theoretical mathematicians tend to be politically liberal, while the practical application mathematicians tend to be more conservative.
In large groups of identification such as African-Americans or evangelicals or progressives, people might adopt at least some of the beliefs because of social identification. This would not conflict with the idea of predisposition to certain modes of thought, as even a slight genetic predisposition could easily lead to associating with similar others, in a powerfully reinforcing social feedback loop. Assortive mating, for example, could easily strengthen belief.
Discussion: I have suggested often that there is a large social, almost tribal, component to liberalism. I suppose it is thus possible that a similar social component exists for conservatism. I think that is more diffuse, however, as there seem to be more kinds of conservatives. But as for group identification, it certainly seems true that there is a subset of conservatives for whom hyphenations of “American” is appalling. One is allowed to identify first with one’s religion, perhaps, but to put any other grouping above American is seen as unpatriotic. To be a hyphenated American is seen as in effect un-American. Worse still would be to see oneself as first a world citizen, or a woman/man. While that view is strongest in one subset of conservatives, it is somewhat true of conservatives in general.
DRJ, commenting over at GM Roper’s, suggests that Democrats are more people-driven, and Republicans more policy-driven. His initial comment:
In reading these and other comments, it strikes me how important specific people are to Democrats as opposed to specific policies. They hate Bush and are thrilled to see Rumsfeld resign. As a conservative, I'm not fond of Pelosi, Rangel, Kennedy, or Dean but I know that others like them are waiting in the wings. It's not the people that bother me near as much as the policies they embrace.
Maybe I'm wrong but it seems like Democrats are more people-driven while Republicans are policy-driven.
When I asked for elaboration, this is part of his longer reply.
My comment was more an observation than a theory although I think there are other factors that tie in. For instance, Democrats are more likely to identify with people based on their personal qualities (such as African-Americans, immigrants, women, etc.) than Republicans, who tend to group according to ideological issues. (e.g., pro-fence immigration reformists, small government/fiscal responsibility proponents, etc.). The Democratic groups are more fixed than the Republican groups, where we tend to realign on issues and make it difficult to stay on the reservation….
I think Democrats are also more likely to be attracted to and motivated by a candidate’s charisma than Republicans, although probably most Democrats think Republicans don’t have charisma with the exception of Ronald Reagan. (I think that’s why Civil Truth views Reagan as both policy-driven and people-driven.) Republicans could field charismatic candidates and Democrats could proffer policy-driven candidates but I don't think those qualities resonate with their respective voting bases.